Wireless Network Failure Detection With Generative Time-Series Alerts

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

5G wireless communication networks face challenges such as limited coverage areas, higher deployment costs, and slow rollout, which can be exacerbated by network function procedure failures, making it difficult to identify and resolve anomalies efficiently.

Innovation Solution

A method using a generative model to interpret time series anomaly detection results for network function procedure failures, incorporating unsupervised multivariate time series anomaly detection and retrieval augmented generation to provide real-time or on-demand interpretations of network failures, reducing delay in troubleshooting and root cause identification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If traditional network monitoring methods are used, then network function procedure failures can be detected, but the time required to identify and resolve anomalies is excessive

Engineering Contradiction:

Improvetime required to identify and resolve anomaliesVSAvoidnetwork performance

Core Design Contradiction:

Loss of timeVSReliability

Solution Approach 1:

The system performs preliminary actions by pre-training the generative model on historical network data and establishing baseline anomaly detection capabilities before actual failures occur. This allows the system to quickly interpret anomalies in real-time without requiring extensive analysis during incident response, thereby reducing the time to identify and resolve network issues while maintaining high reliability

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent introduces a generative model as an intermediary between raw network data and human operators. This intermediary automatically interprets complex anomaly detection results, translates technical data into actionable insights, and provides natural language explanations. This mediation significantly reduces the time required for operators to understand and resolve network failures while ensuring accurate and reliable analysis

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If complex anomaly detection algorithms are implemented, then detection accuracy improves, but system complexity increases

Engineering Contradiction:

Improveanomaly detection accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The generative model serves as an intermediary that simplifies the interface between complex anomaly detection algorithms and end users. It takes the outputs from sophisticated time series analysis and transforms them into interpretable narratives, thereby maintaining high detection precision while reducing the perceived complexity for operators who interact with the system

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent replaces manual analysis mechanisms with an automated generative AI system. Instead of requiring operators to manually analyze complex time series data and anomaly patterns, the system uses machine learning models to automatically detect, interpret, and explain anomalies, thereby maintaining high measurement precision while significantly reducing operational complexity

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If real-time anomaly interpretation is provided, then troubleshooting speed increases, but computational resources required increase

Engineering Contradiction:

Improvetroubleshooting speedVSAvoidcomputational resources

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The system performs preliminary computation by pre-training the generative model on extensive historical network data before deployment. This upfront computational investment enables the model to quickly interpret new anomalies in real-time with reduced computational overhead, thereby increasing troubleshooting speed while managing resource consumption efficiently

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements a hybrid approach where the generative model provides partial real-time interpretation for all anomalies and full detailed interpretation on-demand when operators request additional insights. This partial real-time action maintains high troubleshooting speed for common issues while conserving computational resources by providing excessive detail only when necessary

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS12562965B2Time series transaction failure cause detection and generative alerting for wireless network transactions

Publication Date: 2026.02.24 AT&T INTELLECTUAL PROPERTY I L P

AI summary

A processing system may obtain time series associated with a plurality of network functions of a communication network, each time series including a sequence of values over a plurality of time intervals, each value indicating a percentage of events generated by a respective network function indicating a procedure failure within a respective time interval out of a total number of events generated by the respective network function within the respective time interval, identify, via a time series anomaly detection algorithm implemented by the processing system, an anomaly detection result comprising at least one time interval in which at least one time series exhibits at least one anomaly, apply a query to a generative model requesting an interpretation of the anomaly detection result, where an output comprising the interpretation is generated via the generative model in response to the query, and present the output to at least one endpoint device.